# v12.0.0

Luminous

* Bug #4829: client: handling part of MClientForward incorrectly?
* Feature #11950: Strays enqueued for purge cause MDCache to exceed size limit
* Feature #13025: Add scatter/gather support to librbd C/C++ APIs
* Feature #16523: Assert directory fragmentation is occuring during stress tests
* Bug #16768: multimds: check_rstat assertion failure
* Bug #16807: Crash in handle_slave_rename_prep
* Bug #16886: multimds: kclient hang (?) in tests
* Bug #16914: multimds: pathologically slow deletions in some tests
* Bug #16924: Crash replaying EExport
* Bug #16925: multimds: cfuse (?) hang on fsx.sh workunit
* Bug #17113: MDS EImport crashing with mds/journal.cc: 2929: FAILED assert(mds->sessionmap.get_version() == cmapv)
* Feature #17204: Implement new-style ENOSPC handling in kclient
* Bug #17606: multimds: assertion failure during directory migration
* Bug #17670: multimds: mds entering up:replay and processing down mds aborts
* Bug #17731: MDS stuck in stopping with other rank's strays
* Feature #17834: MDS Balancer overrides
* Feature #17853: More deterministic timing for directory fragmentation
* Feature #17855: Don't evict a slow client if it's the only client
* Bug #17858: Cannot create deep directories when caps contain "path=/somepath"
* Bug #17954: standby-replay daemons can sometimes miss events
* Feature #17980: MDS should reject connections from OSD-blacklisted clients
* Bug #18113: Don't lose deep-scrub information
* Bug #18196: fail to build ceph-11.0.2 if yasm installed
* Bug #18311: Decode errors on backtrace will crash MDS
* Bug #18375: bluestore: bluefs_preextend_wal_files=true is not crash consistent
* Bug #18487: Crash in MDCache::split_dir -- FAILED assert(dir->is_auth()) 
* Bug #18579: Fuse client has "opening" session to nonexistent MDS rank after MDS cluster shrink
* Bug #18600: multimds suite tries to run quota tests against kclient, fails
* Bug #18690: kclient: FAILED assert(0 == "old msgs despite reconnect_seq feature")
* Bug #18743: Scrub considers dirty backtraces to be damaged, puts in damage table even though it repairs
* Bug #18759: multimds suite tries to run norstats tests against kclient
* Bug #18850: Leak in MDCache::handle_dentry_unlink
* Bug #18940: ERROR RESTFUL_IO with S3 GET/PUT operations
* Feature #19075: Extend 'p' mds auth cap to cover quotas and all layout fields
* Feature #19230: Limit MDS deactivation to one at a time
* Bug #19245: Crash in PurgeQueue::_execute_item when deletions happen extremely quickly
* Bug #19253: "failsafe engaged, dropping updates" clog message causing fs suite failures
* Bug #19260: FreesBSD/Clang generates a linking issue about a missing dtor cls::rbd::MirrorImageStatus::~MirrorImageStatus()
* Fix #19288: Remove legacy "mds tell"
* Bug #19407: MgrMonitor doesn't update map when active mgr times out
* Bug #19412: Crash in PyModules::shutdown when non-serve()'ing module is loaded
* Bug #19454: src/arch/ppc.c FreeBSD build break
* Feature #19551: CephFS MDS health messages should be logged in the cluster log